An Integrated Approach to Parameterize Reservoir Models for CO₂ Injection in Aquifers

The injection of carbon dioxide in oil and gas reservoirs (Enhanced Hydrocarbon Recovery, EHR) or in aquifers is a promising way to cope with the short-medium term issue of greenhouse gas emission mitigation. Although CO₂-EHR projects represent the best economical opportunities, the CO₂ injection projects in deep aquifers offer the largest capacity potential for long-term CO₂ storage. The design of a CO₂ injection project mainly depends on the results of reservoir models.

To be predictive, the numerical tools have to be parameterized to adequately model the multiphase flow behavior (relative permeability curves) and the phase exchanges between the injected fluids and the fluids in place (thermodynamic equilibrium). In the particular case of the CO₂ injection, the potential rock/fluid interactions with the host formation must also be studied carefully as a function of the thermodynamic conditions, the fluid compositions, the rock mineralogy but also the flow regime (i.e. the distance to the well bore).

This paper presents an integrated workflow based on experiments and numerical simulations to determine in a comprehensive and robust manner the appropriate parameters at the core scale for a CO₂ injection in aquifer. CO₂ injection experiments were conducted under reservoir conditions. Experiment were carried out on companion plugs under several thermodynamic conditions and with different fluid systems (different brine salinities) in order to collect data (production curves and differential pressure evolution as a function of time) over a wide range of CO₂ storage conditions. Then, these experiments were modeled using a compositional simulator dedicated to CO₂ geological storage.

Using this approach, we show that an unique set of parameters enables modeling of CO₂ injection for various thermodynamic conditions and fluid systems. Therefore, such approach leads to predictive simulations at the core scale, which contributes to the accuracy of the performance prediction of the CO₂ injection at the reservoir scale, if the main geological features (heterogeneities) are included into the large scale model.